scrolling isn't smooth

[iardon]

page scrolling stutters

Page scrolling on forum home, forum display, and show thread all stutter for me. It isn't smooth like it was on 3.8. Other vbulletin sites that are 3.8 don't stutter for me when I scroll their pages.

It's like that on other sites I view that are resource intensive. My computer is actually new but a basic one. Vbulletin should take into consideration those that don't have the top of the line computers.

[monForum]

I cannot re-create this issue in any of my browsers, on any of my different connections.

[Sixthly]

Please post your computer specifications, operating system, website browser, website browser extensions, and running processes. Maybe even tell us how much processor and memory usage is available/free.

[Wayne Luke]

I cannot reproduce this. My computer is two years old as of last month so it isn't state of the art. The HTML is lighter than previously however we cannot ensure that all computers will be able to scroll smoothly.

[iardon]

Please check this thread out... I'm not alone with this issue.

http://www.vbulletin.com/forum/showthread.php?325490-VBulletin-4-forums-laggy

[Kensino]

My computer does the same thing for me. I am using IE8 and System7. I have a brand new machine used to game design so its not because my computer is cheap or old.

It happens when I use the scrollbars or the mouse wheel.

[JonnyThunder]

I've seen similar issues with IE8 on all sorts of sites, not just VB4 based ones. I'd say this is a browser issue.

Try some other browsers - Firefox / Safari etc. And see what it's like with those.

[iardon]

I'm using the latest Firefox... should VB not work correctly with Firefox?

[JonnyThunder]

Yes, I'm also using firefox and it works perfectly on both of my machines. What's the spec of your machine?

Also, have you checked background processes for malware / upgraded mouse driver (for scroll functionality). Does the same happen when you page up and down through the content, or only using scroll wheel?

[ThorstenA]

sorry can not reproduce this.

[iardon]

Vbulletin 3.8 = No scrolling issues
Vbulletin 4 = Scrolling issues

You can ask which browser I'm using or what my computer specs are but the bottom line is I had no problems with previous versions of vbulletin. Also, others have this issue in the thread I posted above.

I'm done with this ticket.

[JonnyThunder]

Okay that may well be true - but it still helps the guys who develop the software to know what kind of conditions people are running the software under. Makes some kind of sense doesn't it?

If someone came to me and told me they had installed Windows 7 and it runs slowly, but they previously had Windows 98 and it worked fine - i'd still ask for specs of the PC. This is a new version of the software and while I agree that you shouldn't have performance issues any more than you had before, perhaps you can help other people to help you by answering the questions they ask (instead of getting arsey about it all)

[AlexanderT]

... perhaps you can help other people to help you by answering the questions they ask (instead of getting arsey about it all)

Arsey? The only person who gets "arsey" here is the one disrespecting those who are trying to help.

Had you bothered to look in the thread posted above, you'd see that there are very concrete examples (along with links to other websites) for what is very likely to cause the behavior described in this bug. Now it's up to vBulletin developers to respond to the new information (so far they haven't).

[JonnyThunder]

Then doesn't it also make sense to provide information surrounding the issue? Such as... oh, I don't know... machine specs, OS version, browser version etc?

It's support 101 - if you have an issue, you post as much information as possible in order to resolve the issue, so that the people who are trying to help can offer some suggestions or the developers can use that information to possibly recreate the environment.

[AlexanderT]

Then doesn't it also make sense to provide information surrounding the issue? Such as... oh, I don't know... machine specs, OS version, browser version etc?

Are you affiliated with vB Solutions? I am asking because you speak in that authoritive voice that seems to indicate such. Otherwise I really don't understand why you don't let a developer of vBulletin handle this issue!


It's support 101 - if you have an issue, you post as much information as possible in order to resolve the issue, so that the people who are trying to help can offer some suggestions or the developers can use that information to possibly recreate the environment.

I hate to repeat myself, but for you I do it anyway: go to that thread linked further above and you will find *very specific* information discovered long time ago by others that shows how Mozilla- as well as Gecko-based browsers (Mozilla more so, though) have a negative influence on CPU usage when a layout uses the experimental CSS3 shadow property (which shows in particular when scrolling a page on slower computers). And also for your information: the current vb4 style makes heavy use of that experimental shadow property, and the more sections in your forumhome or the more posts on a showthread page, the worse the performance penalty.

[JonnyThunder]

Lol.... I like you. You're feisty. Can't we just kiss and make up?

:)

[AlexanderT]

Let's skip the kissing part, but making up sounds good. :)

[JonnyThunder]

Nice one fella. You realise what will happen, is I'll get this problem when my forum goes live...

:) Have a good one!

[msensenbach]

I personally can't reproduce this, but based on the number of users reporting this issue, I'm going to reopen this so potential problems regarding the drop shadow CSS can be investigated.

[Ryan Ashbrook]

I can't reproduce this either.

However, I highly doubt it is caused by the drop shadows, considering people are also reporting issues using IE8, which doesn't support the CSS drop shadow properties.

[Darkimmortal]

I can reproduce this and it is caused by the drop shadows and rounded borders. They come at a cost in Gecko/Windows, just like fixed positioning.

[Ryan Ashbrook]

I can reproduce this and it is caused by the drop shadows and rounded borders. They come at a cost in Gecko/Windows, just like fixed positioning.

How can you prove that is the cause, though? Not trying to start anything, I'm just looking for concrete evidence to prove it. I'm using Firefox 3.5.5 on Windows and scrolling is completely fine for me. And my computer sucks (1.7 GB of RAM and a 2.4 Ghz Pentium 4 running Windows 7). :p

Plus, that doesn't explain why people are experiencing it in Internet Explorer as well.

[AlexanderT]

How can you prove that is the cause, though?

I removed all shadow attributes and immediately saw a huge improvement in terms of scrolling behavior and cpu usage (didn't try removing the rounded corners, it might have had an additional positive effect, too).


I'm using Firefox 3.5.5 on Windows and scrolling is completely fine for me. And my computer sucks (1.7 GB of RAM and a 2.4 Ghz Pentium 4 running Windows 7). :p


Well, maybe it is a matter how to look at it. It is not that scrolling comes to an halt or anything like that; only in relative terms, compared to scrolling pages without these shadow properties is the effect quite visible (have you seen people in the designer communities discussing this everywhere online? Try to google for it). Another way to look at it is to use a task manager on Windows (built-in or Process Manager from Sysinternals) and compare the CPU usage when scrolling a page with and without these shadow properties. Even on my fast dual2core CPU usage increases considerably while scrolling a page with shadow properties up and down compared to a page without shadows. It is also important that you test it with a test forum that offers plenty of section which increases the number of times the browser has to render these shadows for each section on forumhome (my forumhome shows around 50 sections). If it really is necessary, I am willing to make a video snapshot. Btw, the performance penality (with all negative consequences like the relatively inferior scrolling experience) might be a good reason why both mozilla and gecko consider shadow properties support still experimental.


Plus, that doesn't explain why people are experiencing it in Internet Explorer as well.

Let's consider each case separately. IE might suffer from another use/combination of CSS properties in the current vb4 style. CSS is a complicated affair if you want to do it right, and it is yet another reason why it is so important that you'd want a designer with experience who knows these things. For instance, the way you may want to reset default properties (usually at the beginning of a style sheet) can make a huge difference in rendering speed in all browsers. I believe the vb4 beta stylesheet uses a yui-recommended way of resetting properties, which is a good thing (rather than using a global * to reset everything). So the problem might lay somewhere else. This is just an example for a common trap that can lead to negative performance and efficiency which many hobby designers don't know or care about. I haven't looked at all the rules and properties used in the current vb4 beta style sheet, but the list is quite long (some may argue it is bloated in it's current stage), and it is very likely that there are other potential issues, one that might explain the behavior in IE - just like the use of the experimental shadow properties (and probably also the rounded corners properties) explain issues in current mozilla/gecko based browsers.

[Darkimmortal]

I removed all shadow attributes and immediately saw a huge improvement in terms of scrolling behavior and cpu usage (didn't try removing the rounded corners, it might have had an additional positive effect, too).

I can confirm this on Firefox 3.6 beta 2.